Home // DIGITAL 2022, Advances on Societal Digital Transformation // View article

Can "Provably Beneficial AI" Save Us?

Authors:
Selmer Bringsjord
Naveen Sundar Govindarajulu
John Licato

Keywords: machine ethics, robot ethics, inductive reinforcement learning

Abstract:
AI-polymath Stuart Russell, in the face of fear about superhuman AI arriving within 80 years and doing the human race in, commendably offers a recipe (based upon inductive reinforcement learning) for salvation quite different than our own (the sharing of which is beyond the current scope of the present paper). He does this in his recent book *Human Compatible*. Unfortunately, as we explain, Russell's recipe is afflicted by four fatal defects.

Pages: 9 to 12

Copyright: Copyright (c) IARIA, 2022

Publication date: November 13, 2022

Published in: conference

ISBN: 978-1-68558-014-8

Location: Valencia, Spain

Dates: from November 13, 2022 to November 17, 2022